Improving machine learning methods for solving non-stationary conditions based on data availability, time urgency, and types of change

• Main Author: Goh, Chun Fan
• Other Authors: Seet Gim Lee, Gerald
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Mechanical engineering::Robots; Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence
• Online Access: https://hdl.handle.net/10356/147041
• Institution: Nanyang Technological University

Supervised learning algorithms do not work well when the deployment condition is dissimilar to the training condition. Such non-stationary conditions include covariate shifts and concept shifts. Importance weighted learning (IWL) is used to handle a one-time covariate shift but not frequent shifts a...